Package LBJ2.infer

Inference algorithms are implemented here (derived from Inference), but most of the classes in this package are used internally by LBJ at runtime to represent constraints and to translate between constraint representations.

See:
          Description

Interface Summary
ILPSolver Classes that implement this interface contain implementations of algorithms that solve Integer Linear Programming problems.
 

Class Summary
ArgumentReplacer Anonymous inner classes extending this class are instantiated by the code generated by the LBJ compiler when creating FirstOrderConstraint representations.
AtLeastQuantifier An "at least" quantifier states that the constraint must hold for at least m of the objects in the collection.
AtMostQuantifier An "at most" quantifier states that the constraint must hold for no more than m of the objects in the collection.
Constraint A constraint is an expression that is either satisified or unsatisfied by its constituent classifier applications.
EqualityArgumentReplacer Anonymous inner classes extending this class are instantiated by the code generated by the LBJ compiler when creating FirstOrderConstraint representations.
ExistentialQuantifier An existential quantifier states that the constraint must hold for at least one object from the collection.
FirstOrderBinaryConstraint Represents a first order constraint involving a binary operator.
FirstOrderConjunction Represents the conjunction of first order constraints.
FirstOrderConstant A first order constant is either true or false.
FirstOrderConstraint All classes for representing first order constraints are derived from this base class.
FirstOrderDisjunction Represents the disjunction of first order constraints.
FirstOrderDoubleImplication Represents a double implication between two first order constraints.
FirstOrderEquality Represents either an equality or an inequality between two values, a classifier application and a value, or two classifier applications.
FirstOrderEqualityTwoValues Represents the comparison of two String values.
FirstOrderEqualityWithValue Represents the comparison of a classifier application with a value.
FirstOrderEqualityWithVariable Represents the comparison of two classifier applications.
FirstOrderImplication Represents an implication between two first order constraints.
FirstOrderNAryConstraint Represents a first order constraint with an arbitrary number of arguments, usually assumed to be greater than or equal to 2.
FirstOrderNegation Represents the negation operator applied to a first order constraint.
FirstOrderVariable Represents a classifier application.
GLPK Deprecated. As of LBJ release 2.0.12, it is preferrable to pass a GLPKHook object to the ILPInference constructor.
ILPInference This class employs an ILPSolver to solve a constrained inference problem.
Inference An object of this class keeps track of all the information necessary to perform inference.
InferenceManager The inference manager is a cache of Inference objects accessed via their names and head objects.
InvocationArgumentReplacer Anonymous inner classes extending this class are instantiated by the code generated by the LBJ compiler when creating QuantifiedConstraintInvocation representations.
NaiveGLPK Deprecated. As of LBJ release 2.0.12, it is preferrable to pass a GLPKHook object to the ILPInference constructor.
ParameterizedConstraint This class represents an LBJ constraint as it appears in a source file.
PropositionalAtLeast Represents the constraint that at least m of the children constraints must be true.
PropositionalBinaryConstraint Represents a propositional constraint involving a binary operator.
PropositionalConjunction Represents the conjunction of two propositional constraints.
PropositionalConstant A propositional constant is either true or false.
PropositionalConstraint All classes for representing propositional constraints are derived from this base class.
PropositionalDisjunction Represents the disjunction of two propositional constraints.
PropositionalDoubleImplication Represents a double implication between two propositional constraints.
PropositionalImplication Represents an implication between two propositional constraints.
PropositionalNAryConstraint Represents a propositional constraint with an arbitrary number of arguments, usually assumed to be greater than or equal to 2.
PropositionalNegation Represents the negation operator applied to a propositional constraint.
PropositionalVariable Every propositional variable is Boolean and represents one possible prediction from a classifier application.
QuantifiedConstraintInvocation Represents the invocation of a parameterized constraint nested inside at least one quantification expression, where the parameter is a function of the quantification variables.
Quantifier A quantifier is a first order constraint parameterized by an object taken from a Java Collection of objects.
QuantifierArgumentReplacer Anonymous inner classes extending this class are instantiated by the code generated by the LBJ compiler when creating FirstOrderConstraint representations.
UniversalQuantifier A universal quantifier states that the constraint must hold for all objects from the collection.
XpressMPHook This interface to Xpress-MP from Dash Optimization is designed to work with Xpress-Optimizer 15.25.02.
XpressMPHook.Vector2D A two dimensional vector representation implemented here because we are not assuming Java 5.
XpressMPHook.Vector2DDouble A two dimensional vector representation implemented here because we are not assuming Java 5.
XpressMPHook.Vector2DInt A two dimensional vector representation implemented here because we are not assuming Java 5.
 

Exception Summary
InferenceNotOptimalException Exceptions of this type are thrown by the ILPInference class when the selected ILPSolver did not successfully find the optimal solution to the inference problem.
 

Package LBJ2.infer Description

Inference algorithms are implemented here (derived from Inference), but most of the classes in this package are used internally by LBJ at runtime to represent constraints and to translate between constraint representations.